Modeling and forecasting petroleum futures volatility

Forecasts of oil price volatility are important inputs into macroeconometric models, financial market risk assessment calculations like value at risk, and option pricing formulas for futures contracts. This paper uses several different univariate and multivariate statistical models to estimate forecasts of daily volatility in petroleum futures price returns. The out-of-sample forecasts are evaluated using forecast accuracy tests and market timing tests. The TGARCH model fits well for heating oil and natural gas volatility and the GARCH model fits well for crude oil and unleaded gasoline volatility. Simple moving average models seem to fit well in some cases provided the correct order is chosen. Despite the increased complexity, models like state space, vector autoregression and bivariate GARCH do not perform as well as the single equation GARCH model. Most models out perform a random walk and there is evidence of market timing. Parametric and non-parametric value at risk measures are calculated and compared. Non-parametric models outperform the parametric models in terms of number of exceedences in backtests. These results are useful for anyone needing forecasts of petroleum futures volatility.

[1]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[2]  Richard G. Harris,et al.  Applied Time Series Modelling and Forecasting , 2003 .

[3]  Peter C. Fusaro.  Energy risk management , 1998 .

[4]  Ronald W. Masulis,et al.  Energy Shocks and Financial Markets , 1996 .

[5]  R. Engle.  Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[6]  P. Franses,et al.  Forecasting stock market volatility using (non‐linear) Garch models , 1996 .

[7]  T. Bollerslev,et al.  Generalized autoregressive conditional heteroskedasticity , 1986 .

[8]  Adrian Pagan,et al.  Alternative Models for Conditional Stock Volatility , 1989 .

[9]  Heather Mitchell,et al.  Generalized asymmetric power ARCH modelling of exchange rate volatility , 2002 .

[10]  Nigel Meade,et al.  A comparison of the accuracy of short term foreign exchange forecasting methods , 2002 .

[11]  J. Peter Ferderer,et al.  Oil price volatility and the macroeconomy , 1996 .

[12]  L. Glosten,et al.  On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks , 1993 .

[13]  Perry Sadorsky.  Oil price shocks and stock market activity , 1999 .

[14]  V. Akgiray.  Conditional Heteroscedasticity in Time Series of Stock Returns: Evidence and Forecasts , 1989 .

[15]  W. Fuller,et al.  Distribution of the Estimators for Autoregressive Time Series with a Unit Root , 1979 .

[16]  James D. Hamilton.  Time Series Analysis , 1994 .

[17]  Pierre Giot,et al.  Market risk in commodity markets: a VaR approach , 2003 .

[18]  Francis X. Diebold,et al.  Elements of Forecasting , 1997 .

[19]  Anil K. Bera,et al.  Efficient tests for normality, homoscedasticity and serial independence of regression residuals , 1980 .

[20]  Chris Brooks,et al.  Volatility forecasting for risk management , 2003 .

[21]  R. Chou,et al.  ARCH modeling in finance: A review of the theory and empirical evidence , 1992 .

[22]  J. Zakoian.  Threshold heteroskedastic models , 1994 .

[23]  Carol Alexander,et al.  Market Models: A Guide to Financial Data Analysis , 2001 .

[24]  Philippe Artzner,et al.  Coherent Measures of Risk , 1999 .

[25]  R. Cumby,et al.  Testing for market timing ability: A framework for forecast evaluation , 1987 .

[26]  James G. MacKinnon,et al.  Numerical Distribution Functions for Unit Root and Cointegration Tests , 1996 .

[27]  W. Newey,et al.  A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelationconsistent Covariance Matrix , 1986 .

[28]  P. Phillips.  Testing for a Unit Root in Time Series Regression , 1988 .

[29]  J. David Cabedo,et al.  Estimating oil price ‘Value at Risk’ using the historical simulation approach , 2003 .

[30]  Chris Brooks,et al.  Model Choice and Value-at-Risk Performance , 2002 .

[31]  R. Engle,et al.  Multivariate Simultaneous Generalized ARCH , 1995, Econometric Theory.

[32]  T. Brailsford,et al.  An evaluation of volatility forecasting techniques , 1996 .

[33]  Perry Sadorsky.  The macroeconomic determinants of technology stock price volatility , 2003 .

[34]  Russell P. Robins,et al.  Estimating Time Varying Risk Premia in the Term Structure: The Arch-M Model , 1987 .

[35]  Kiseok Lee,et al.  Oil Shocks and the Macroeconomy: The Role of Price Variability* , 1995 .

[36]  John Moody,et al.  Developments in forecast combination and portfolio choice , 2001 .

[37]  William Thomson,et al.  Two characterizations of the Raiffa solution , 1980 .

[38]  L. Glosten,et al.  Economic Significance of Predictable Variations in Stock Index Returns , 1989 .

[39]  W. Newey,et al.  A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelationconsistent Covariance Matrix , 1986 .

[40]  Philippe Jorion,et al.  Risk2: Measuring the Risk in Value at Risk , 1996 .

[41]  P. Bossaerts,et al.  Implementing Statistical Criteria to Select Return Forecasting Models: What Do We Learn? , 1999 .

[42]  Philippe Jorion.  Value at risk: the new benchmark for controlling market risk , 1996 .

[43]  Jun Yu,et al.  Forecasting volatility in the New Zealand stock market , 2002 .

[44]  K. West,et al.  The Predictive Ability of Several Models of Exchange Rate Volatility , 1994 .

[45]  Mike K. P. So,et al.  Forecasting exchange rate volatility using autoregressive random variance model , 1999 .